<p>welcome, catmom - you sound like a great cataunt too. You might want to reference some the actual results on the HYM forums for your DN - if he’s a math kid, he’ll realize the implications, even if his mom is still wearing blinders at this point.</p>
<p>also welcome, shoshi. The advantage of keemum’s formula is that one could set up a spreadsheet, and do the calculations on an individual basis. For example, UPenn seems to like legacies if they apply ED.</p>
<p>fwiw re legacy:
at Stanford info session, adcom gave this info (I am a legacy): Legacy = Stanford degree, whether undergrad or graduate; all considered the same. Legacy status admit history indicates that legacy has statistically doubled the chances of acceptance. It is NOT that they give you twice the “points” or anything specific like that, but this is what they have observed re outcomes.</p>
<p>Of course, “doubling” your chances from 90% reject to 80% rejects is still low chance/lottery.</p>
<p>So, at this school, you could use the data to defend either position: doubles your chances and “doesn’t mean much,”</p>
<p>catmom, Ohio-Mom is right. There’s a ton of hard AND anecdotal information on this site that should be brought to the attention of your SIL and DN. It’s a sad day when the blinders come off and parents/kids realize that not everybody who is well-qualified for HYM gets accepted. But far better to have those blinders off now as opposed to next April.</p>
<p>“afan: I’m not sure how well that legacy thing works anymore.”</p>
<p>I certainly would not count on it. </p>
<p>I agree that one should avoid overestimating chances of admission. My point is that the formula fails on the simple statistical grounds of assuming independence when this is clearly not the case. It also assumes that it is possible to know whether one is “typical” based on SAT score alone. As others have noted, at a place where 25% of students have SAT’s below 1400, many of those below 1500 were admitted IN SPITE of “low” test scores. They had something else important that made them stand out, even if they had, what were to Harvard, below average SAT’s.</p>
<p>There is a better approach than attempting spurious precision in estimates of admission prospects. Get over the idea that only HYP can provide a good education and a satisfactory college experience. The students who are reasonable candidates for these schools should be looking at a group that includes, for example, the USNews top 30 LAC and top 30 universities. They are all terrific. For someone who is not a Caltech Axline scholar, any of this group of 60 could provide a first rate education. The way to avoid getting rejected by all is to avoid applying only to lottery ticket schools.</p>
<p>This is not an endorsement of USNews, or its ridiculous rankings. I chose as a way of indicating the number of top schools out there.</p>
<p>“My point is that the formula fails on the simple statistical grounds of assuming independence when this is clearly not the case.”</p>
<p>Unless you (honestly) complete the part on the application that asks where else you applied, applying to Case is independant of applying to Colby is independant of applying to Chicago. If my son chooses not to apply to Colby, it does not, in and of itself, affect his chances at Case and Chicago in the slightest. This is the statistical independance we are talking about - not if its a lousy application or a good application. </p>
<p>If you are spreading yourself too thin and sending, say, 15 mediocre apps instead of 6 great ones, you will likely hurt your chances everywhere. But this is a different concept.</p>
<p>Now, there was that time a few years ago with, correct me if I am wrong, Princeton hacked Yales website (or vice versa) when there was collusion and applications were not independant…</p>
<p>
This misses the meaning of statistical independence. According to the definition on wikpedia, for example, “when we assert that two random variables are independent, we intuitively mean that knowing something about the value of one of them does not yield any information about the value of the other.”</p>
<p>The formula in question assumes that the chances of admission at each college are these random variables. Then in figuring the probability of X not occuring, you simply multipy the chances for each independent part of X not occuring. But is this the case for college admissions? If you believe that, then you have to believe that either the adcoms actually are rolling dice or that the criteria used by any one college is not replicated in the slightest degree by any other college. I doubt the former, and I think few people subscribe to the latter. The correlation may not be 100%, but it is quite far from zero.</p>
<p>Its pretty well known that selective colleges consider ECs, GPA, test scores, recs, essays, and so on. Maybe they won’t assign the same weights to each, and some factors are evaluated subjectively. But an essay that stinks at Harvard is unlikely to win huzzah’s at Yale, a rec seen as lukewarm at Amherst is not going to be seen as enthusiastic at Williams.</p>
<p>To say that “If my son chooses not to apply to Colby, it does not, in and of itself, affect his chances at Case and Chicago in the slightest. This is the statistical independance we are talking about - not if its a lousy application or a good application.” is to adopt your own private meaning for the term “statistical independence”. I don’t see how doing so helps make anything clearer. A good or lousy application is exactly what we are talking about, and they are likely to be evaluated similarly at the schools receiving them.</p>
<p>mikemac -
my defination is from my stats courses in grad school, not wikipedia.</p>
<p>Well I will be polite, but ask you to provide a definition of statistical independence that supports the way you are using it. Lots of us took stats in grad school, and no one else uses the term, or the formula, as you are.</p>
<p>Mikemac is correct that for the formula to be applicable, the probability of admission to school A must be uncorrelated with the probability of school B. In this case, if School A uses SAT, EC, GPA, recommendations, then School B must use eye color, hat size, and square root of birthdate day of month, or some other set of considerations that have nothing to do with those of school A. If School A and B use common factors, then they are not independent, and the formula just does not work.</p>
<p>Did your stats courses cover computing conditional probabilities for correlated events?</p>
<p>I’m in complete agreement with afan and mikemac on this issue, as is consistent with my first comment on this thread concerning independence. </p>
<p>Admission isn’t a lottery, nor is applying like buying a lottery ticket. Your odds of admission depend on your individual characteristics (SAT, GPA, race/ethnicity, where you live, etc.), and those characteristics are likely to determine your odds of winning (admission) at virtually all of the schools to which you apply. Therefore a decision at one is likely to be correlated with a decision at the others – assuming there is at least some selectivity based on individual qualifications (not just paying the application fee) at all the schools.</p>
<p>someone <em>must</em> have already stated this but this might not be so obvious: rates of acceptance do not signify probability. They’re just statistics. </p>
<p>It should also be mentioned that college decisions are independent events, meaning, you cannot simply multiply your “probabilities” together. </p>
<p>The formula is flawed mathematically. Also, reducing your college admission process to a simple formula is so inhuman. View the process as something that you have control over–not something that “math” can dictate.</p>
<p>mikemac has it right. The outcomes are not independent. Like schools think alike to a large degree. Further, some of them will be reading the same essays, even if slightly reworked.</p>
<p>It would be interesting (well, a little) to have the data to do a study on the collective Ivy apps regarding multiple admissions and rejections.</p>
<p>Edit: the acceptance rate IS the empirical probabilty of a random group of applicants. Multiplying is the correct technique</p>
<p>You’re right that the outcomes are not truly independent. On the other hand, there seems to be little consistency in the admissions decisions process. If the decisions were well correlated, we would see more students accepted to all their Ivy schools and more students rejected by all of them. Instead we see students who are rejected by all the Ivies except one. Or they’re rejected by 2 Ivies and accepted by 2. Or they’re rejected by their matches but accepted to their reaches.</p>
<p>If you visit the college stats websites and examine the statistics of applicants to the most selective schools, the stats look remarkably similar. Everyone has SAT I scores >700. Everyone has SAT II > 700. Everyone has several AP scores of 5. Everyone has a 4.0 GPA. . Look up a school and examine the stats for the accepted students. Now examine the stats for the rejected students. You’ll see accepted students with outstanding stats and rejected students with mediocre stats. You’ll also see rejected students with outstanding stats and accepted students with mediocre stats.</p>
<p>Assuming the applicant is unhooked, this means admissions decisions are based primarily on XCs and essays. That makes the decision process much more unpredictable, much more subjective, much more like a lottery. </p>
<p>I’ve looked at samples of successful Ivy League college application essays. Surprisingly, the vast majority of them are unimpressive. A very few are outstanding. I’ve also read unsuccessful Ivy League college application essays. They look amazingly like the successful ones. If you’re lucky, the adcom will find something appealing about your essay, some indication that you would be a fine addition to their campus. Or you might be unlucky.</p>
<p>We’re left with the XCs. It’s true that some applicants have exceptional XCs like Olympic medalist or CD recording with the Chicago Symphony but they are very rare. Again, visit the college stats sites and you will find that the vast majority of applicants to selective schools have impressive but not earthshattering XCs. How does the adcom decide? How do they choose between the basket weaving expert and the pogo stick record holder? It’s pretty subjective, I think.</p>
<p>Perhaps the outcomes are less independent at the less selective schools. At the most selective schools, however, it’s a different story. A small percentage of the applicants are shoo-ins, a small percentage are automatic rejects. The vast majority of applicants have resumes that look almost exactly alike. That makes the decision process a fairly random one. Like a lottery.</p>
<p>Ok, pure math people, please provide your model.</p>
<p>Keemun, it’s only random in the sense that an outsider cannot predict the outcome. Inside the U, it’s not all that random. </p>
<p>At a school where there are several large, significant publications to be staffed, the Adcoms know how many aspiring journalists need to be admitted every year to keep that machinery going. If there are three chamber orchestras, two wind ensembles, and a symphony orchestra and no music conservatory, they know how many cellists and tuba players they have to identify among the piles of math and history majors. And so on.</p>
<p>To you it looks random-- the Val from your HS gets in; the Val from mine does not, both have scores in the same range. But given the need of the university, my tennis playing Val may not fit the needs of the college as well as your Tuba playing Val. That’s hardly random. Kids who fit a need get admitted at a higher rate than kids who don’t.</p>
<p>I think the lottery analogy continues to feed the frenzy of admissions since it implies two things-- 1) you gotta be in it to win it, which encourages thousands of kids without a snowball’s chance in hell to buy the Princeton sweatshirt and pop in an application, 2) The more tickets you hold, the greater the odds. I wish that Princeton wearing sweatshirt kid would spend time at BC and CT College and other places that are similar to Princeton in many ways rather than spending their time applying to Columbia and Brown since “it’s a lottery and I want as many tickets as I can buy”. But-- that’s me, and as I’ve said before on this and other threads, I don’t buy into this deterministic, formulaic view on college admissions anyway. Do your kid a favor-- invest time helping identify what makes him/her special and fabulous, and then go explore colleges that will 1- admit him/her, and 2- build on that fabulousness. The families I see in distress are those that spend time gaming the odds and then come April are looking at one or two dismal choices which don’t match the kid one iota, or which they can’t afford without sacrificing their retirement, and then they want to know why “The admissions game” is so unfair.</p>
<p>Blossom - </p>
<p>“To you it looks random-- the Val from your HS gets in; the Val from mine does not, both have scores in the same range. But given the need of the university, my tennis playing Val may not fit the needs of the college as well as your Tuba playing Val. That’s hardly random. Kids who fit a need get admitted at a higher rate than kids who don’t.”</p>
<p>Exactly: it looks random to us on the outside. The more research you do (contacting the Tennis coach and the musical director for brass, for example) the better chance you have gleaning some knowledge as to whether the school might want you, and whether you might want the school. Of course, neither you nor the school know how may Tubas or Tennises will apply in a given year, but at least its something.</p>
<p>“at Stanford info session, adcom gave this info (I am a legacy): Legacy = Stanford degree, whether undergrad or graduate; all considered the same. Legacy status admit history indicates that legacy has statistically doubled the chances of acceptance. It is NOT that they give you twice the “points” or anything specific like that, but this is what they have observed re outcomes.”</p>
<p>Statistically, an income over $150k multiplies one’s chances of admissions at HYPS, from what I can tell, roughly six times over that of applicants from family incomes between $45k-$150k.</p>
<p>The odds of that happening by lottery quickly approach zero.</p>
<p>mini,
last fall, I tried to get a handle on how many unhooked boys from Ohio might be offered admission to H. I think think I came up with 3, and maybe that was high. My son decided - not worth the effort.</p>
<p>blossom, I was using the word “random” in the mathematical sense, not in the traditional sense, because I was addressing the independent outcomes question. The fact that every college has different holes to fill supports my view that acceptance decisions are mostly independent, that an advantage at one school does not necessarily imply an advantage at another.</p>
<p>I agree that the college admissions scene has become a ridiculously frantic affair. Most students would be better off avoiding the most selective colleges altogether. For a small percentage of applicants, though, the most selective colleges may be the best matches. My own experience with schools has run the gamut from ghetto school to neighborhood school to experimental school to magnet school to top LAC to top Ivy. Each had its advantages and disadvantages. I am under no illusion that the Ivies are better in every way, that they are an automatic ticket to success. What the Ivies provide is access to the leading experts in almost every academic discipline. Some of these experts may be inept teachers but virtually all have in-depth knowledge about their field of research. </p>
<p>For those interested in scientific research, the prestigious schools also offer state-of-the-art labs. I just read the biographies of two recent Nobel Prize winners. Both started their research as undergraduates, one at Stanford and one at MIT. The state-of-the-art equipment they had access to is simply not available at LACs and most state univs. If these two scientists had not attended these selective schools, they probably would not have made the scientific discovery that led to their Nobel prizes.</p>
<p>“they probably would not have made the scientific discovery that led to their Nobel prize”</p>
<p>Harold Varmus is a member of the National Academy of Science, winner of a Lasker award and a Nobel prize. He served as head of the National Institutes of Health and is currently the head of the Sloan Kettering cancer center. </p>
<p>Dr. Varmus recieved his undergraduate degree from Amherst, where he was an English major.</p>
<p>He shared his Nobel prize with his collaborator, Michael Bishop, who was a chemistry major… at Gettysburg College.</p>
<p>My point was not that you need to attend an Ivy to win a Nobel. There are plenty of non-Ivy grads who have won Nobels. I was addressing blossom’s view that no one should should bother with the most selective colleges. My point was there is a small percentage of applicants for whom selective schools like Amherst, Stanford and MIT are best matches. Because it’s so difficult to be admitted, they are forced to examine the admissions process carefully and think about how best to improve their chances of acceptance.</p>
<p>With regard to scientific research, some areas of research can be done at a LAC while other areas require special, often very expensive, equipment available at just a few U.S. institutions. Because the most selective schools generally have the largest endowments, the most expensive equipment will often be found there.</p>
<p>Some of my friends loved the Ivies they attended and others hated them. Ivies will not suit everyone. Most students would be better off attending a LAC like I did because the teaching at LACs is generally better. LACs offer smaller classes and professors who regard teaching, rather than research, as their main priority. For academic standouts, though, some of the most selective schools may be best matches so I would encourage them to apply but not to expect to be admitted.</p>